# The ‘FunChisq’ R package

### Overview

The package provides statistical hypothesis testing methods for inferring model-free functional dependency. Functional test statistics are asymmetric and functionally optimal, unique from other related statistics. The test significance is based on either asymptotic chi-squared or exact distributions.

The tests include asymptotic *functional chi-squared tests* (Zhang & Song, 2013) <arXiv:1311.2707> and an *exact functional test* (Zhong & Song, 2019) <10.1109/TCBB.2018.2809743> (Nguyen et al., 2020) <10.24963/ijcai.2020/372>. The *normalized* functional chi-squared test was used by Best Performer NMSUSongLab in HPN-DREAM (DREAM8) Breast Cancer Network Inference Challenges (Hill et al., 2016) <10.1038/nmeth.3773>.

To measure the effect size, one can use the asymmetric *function index* (Zhong & Song, 2019) <10.1186/s12920-019-0565-9> (Kumar et al., 2018) <10.1109/BIBM.2018.8621502>. Its value is minimized to 0 by perfectly independent patterns and maximized to 1 by perfect non-constant functions.

A simulator (Sharma et al., 2017) <10.32614/RJ-2017-053> can generate functional, non-functional, and independent patterns as contingency tables. The simulator provides options to control row and column marginal distributions and the noise level.

### When to use the package

Tests in this package can be used to reveal evidence for causality based on the causality-by-functionality principle. They target model-free inference without assuming a parametric model. For continuous data, these tests offer an advantage over regression analysis when a parametric functional form cannot be assumed. Data can be first discretized, e.g., by R packages ‘Ckmeans.1d.dp’ or ‘GridOnClusters’. For categorical data, they provide a novel means to assess directional dependency not possible with symmetrical Pearson’s chi-squared or Fisher’s exact tests. They are a better alternative to conditional entropy in many aspects.

### To download and install the package

`install.packages("FunChisq")`